Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
German
whisper
Generated from Trainer
Instructions to use Sandiago21/whisper-large-v2-german-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Sandiago21/whisper-large-v2-german-2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Sandiago21/whisper-large-v2-german-2")# Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("Sandiago21/whisper-large-v2-german-2") model = AutoModelForSpeechSeq2Seq.from_pretrained("Sandiago21/whisper-large-v2-german-2") - Notebooks
- Google Colab
- Kaggle
Conversion to GGML for usage with whisper.cpp
#3
by Reknezd - opened
I want to use this fine-tuned model with whisper.cpp and therefore converted it to GGML with the following Python script:
https://github.com/ggerganov/whisper.cpp/blob/master/models/convert-h5-to-ggml.py
However, the GGML model generated in this way produces completely unusable results. This might be caused by a problem with the conversion, which is why I’m asking: has anyone had similar experiences converting this model and knows of a solution?